By request -- Tutorial: the easiest possible Photoshop landscape painting

All illustrations are uploaded
at full size -- you'll
need to view them at 1:1
to see them properly!

If you have a pen tablet and the skill to draw from scratch, any picture will start with a sketch. It’s the “proper” way to create a picture … but not too many people can actually draw! Even artists who can draw invariably start with photo references. The finished picture will, in almost every instance, be an amalgam of two or several photographs, where you have traced and combined desired elements from each and discarded the unwanted features.

Now, if you can draw, Step One is to start the project with at least a sharp idea in mind of what you’re trying to create. Once you have the image firm in the mind’s eye, Step Two is to gather the reference photos, get into your paint program (Photoshop is great for this … so is Krita), and start tracing off the individual elements that will go together to make your unique image.

But this little tutorial is going to assume you can’t draw –- or that you don’t have a pen tablet, and no immediate plans to buy one. There’s a hundred reasons for not having one, and not planning to buy one. This doesn’t mean you can’t create lovely images, using tools that are at your disposal, if you have any paint program, or Photoshop (yes, even Elements).

So, assuming you can’t draw from scratch, any image you create is always going to start with one or more photos … but rather than tracing off elements from them and combining them into a fresh sketch, here. we’re going to start simple. In this tutorial, we’ll take the simpler (faster) route, and create a painting that takes no drawing at all. None. I promise.

It all starts with at least one photograph! In a perfect world, it would be your own snapshot, but if you live in a desert and want to paint the ocean, or on the plains, and you want to paint mountains, you’ll have to find your base photograph(s) elsewhere. Ideally, these need to be free images. You don’t want to actually steal photos from photographers ― luckily, you don’t have to.

The Internet is teeming with freebie sites. There’s everything from the free wallpaper sites, which supply high quality images, to user-driven sites like pixabay.com … millions of free images that are posted by users via another form of visual social media. They’ve been made available for visitors. There is also Wikimedia, brim-full of non-copyright images. As a last resort, consider the actual gist of copyright law: a new work (literary or artistic) must be 15% different from the original that was its source. Well, even supposing you stole a photo from a photographer, the image you’ll be creating will be about 75% different, so if push came to shove … well, I leave the ethics of the art to you! I’m not here to talk ethics, rights and royalties!

Found on Pixabay (free) ... it'll do for the base image!

So, this will be my picture. I’ve sized it to 1600 to 1067. Why? Two reasons. One: this is one of the standard aspect ratios for images. Two: it’s not very big, because it doesn’t have to be very big … because I’m not painting by hand on this one. If I were going to paint with the pen tablet, using Photoshop or Krita, I would set the size to at least 2500 pixels wide. You need the wiggle room to paint ― but there is a major downside to working with colossal images. Unless you have a supercomputer, the processor can struggle when the images get so big. Any hesitation, lag or delay between what your virtual brush is doing and what appears on-screen can be fatal when you’re painting. Rather paint on a smaller canvas and avoid strife. You’ll know the limitations or otherwise of your processor, so sizing the image at this stage is up to you.

Next, you’ll notice that the picture I’ve chosen is extremely flat. This means it has low contrast, no vivid colors, no real white or black zones, and a lot of chromatic grays. This is what you want, to start with. A flat picture will accept the abuse you’re going to heap onto it, when you come to apply all the filters, all that dodging and burning, the merging of layers with blend modes, the over-burning of the lens flare you’ll be adding. If you start out with a high-contrast image, it’ll be stark before you know it, mostly black and white. So, if your image is not soft and flat to begin with ― go into Photoshop (or whatever), and deliberately make it flat and soft.

(This tutorial assumes that you know how to do this, in whatever paint or photo program you’re using. In Photoshop, for instance, you want to go Enhance > Adjust Lighting > Levels … and just play around till you see what you want. If you don’t know how to do this, just play. Make a mess, have fun … it’s the best way to learn!)

So, now we have our base picture. We’ll put it into Layer One, and rename the layer ORIGINAL PHOTO. Notice all those other layers above it, which I’ve turned off in order to show the base picture. Those layers are where the magic happens, and you’ll be in charge of every filter, blend and merge. Starting with the same image, ten artists will come up with ten different pictures, barely alike. It’s your work … you choose what’s happening on the canvas.

TIP: when you do anything in a project ― apply a filter, adjust the lighting or sharpness, always, always DUPLICATE LAYER before you move on and do one more thing. Work on the clone layer, the copy, not the original layer. If you do further work on a clone, and you make a mess (it’s inevitable; it will happen; no one is immune to this rule!), you can throw the mess in the bin and go back to the original layer. Clone it again before doing things to it. This will save your sanity, if not your life!

If you were going to actually draw now, using the base image as your source, you would start your pen tablet (Wacom or Huion, or whatever; I have used both, and currently use the Huion while the Wacom Bamboo is put away). You would create a new layer op top of the photo … you would select a brush to draw/paint with … pick a color from the photo to paint in … and, in the new layer, you’d start to trace over the shapes and forms of the trees, plants, water and ground. You’d change and adapt what you wanted to change ― say, take out a fencepost, or get rid of a parked car, or omit the trash somebody dumped!

Since you’re not drawing here, instead, we’re going to use a shortcut. Photoshop has a wide variety of filters that can sometimes give a photo some of the characteristics of art. The filters are not fool proof. They often do not work well. Even when they do work, if you just slap a filter onto a photo and call it “done,” all you have is a Photoshop filter slapped onto a photograph … it doesn’t look very good. It certainly isn’t an enchanting painting.

But the process of adding a filter to the photo is the first step in getting to the painting. This is the trick that saves you having to draw, if you can’t draw, don’t have a tablet, or don’t have the time. There are all kinds of filters, some quite good, some really bad, some utterly pointless. A handful can work well some of the time, but don’t expect them to work every time ― it can be a bit of a crapshoot!

It’s up to you to choose the filter you like the best. No one can choose this for you … you are the artist, and this is your project. The photo is going to turn into a painting in your way. No one else’s!

You’ll find the filters here: Filters > Filter Galley … and you’ll need to sort the wheat from the chaff. It’s mostly chaff. Very few of these filters will do anything at all for you. If you apply just one, the result will probably be boring, lifeless, flat ― nothing enchanting about it. This is just the first step.

So, here is Step Three: apply the first filter, whatever you choose. In my picture, I applied Dry Brush, with a brush size of 0, brush detail of 10, and texture of 3. You could use anything you like … but remember that what you choose must be matched to what the picture needs, or it won’t work out very well. This is why “Zen Photoshop” really is art, not just a mechanical process.

Experiment with all the filters. Apply them all, configure them all, delete what you don’t like till you get to something you do like. Save the file. Now that you have Step Three to your liking, immediately duplicate the layer … you’re going to be working on the clone, not the original!

Okay, now we’ve got some lines into the picture, which simulates the work we’d have done by sketching or tracing, the next thing to think about is the color scheme, or pallet. If I’d sketched the whole thing by hand, color would still have been the next thing to tackle; at this point, the process is fairly similar. Now, if I were painting by hand with the pen tablet, this is the moment where I’d take a large, soft brush and start to put in swathes of color, which set the tone, and would also decide which will be the light and dark areas in the finished piece…

Using the shortcut method, it’s not so different. To a certain extent, the colors are already in the picture, because it started as a photo ― but the colors are all wrong. I had an idea in mind when I started, and this picture is not even close. I wanted “early morning in an enchanted forest,” but the photo I started with is just a kind of misty, murky twilight. All it gives me are the shapes of the trees. That’s only the starting point. I need to get the colors I want, or it’s never going to look like an enchanted woodland.

There’s two ways to go. You might get lucky with Photoshop’s “colorize image” function ― I haven’t had much luck with it, and seldom use it, but it’s worth trying. To do this, convert your cloned layer to monochrome, and then … colorize it. See what happens. Don’t be surprised if it’s not very good; it seldom is!

So, I always go the other way, and deliberately set the specific color pallet I want. Okay, so how do you do this? It’s actually not that difficult. In your mind’s eye, see the colors you want. Now, open the color picker pallet. Choose two colors that represent the most opposite colors in your imagined painting ― it could be the sky and the ground, or the light and the shadow areas. Set those two colors as the active swatches ― ie., they will be the ones your brushes are painting with, and which are used by the bucket fill tool … and when you option the gradient tool, you’ll get both.

This is what you’ll be doing. With the colors picked according to the image in your mind’s eye, select gradient fill. Create a new layer ― and fill it with a gradient. It’s hit and miss in Photoshop, so keep stroking the input path across the screen till you see what you want. Got it?

So far so good. Now, put a blend mode onto the layer to bake those colors into the layer below, which has the photo converted to lines. Aha! The magic begins as the picture assumes the colors you want! For my picture, the blend mode I used was Overlay, at 100% opacity. You could choose from half a dozen blend modes, and adjust the opacity from 10% to 100%. Only you know what you want your picture to look like. Get in there and play with the blend modes and opacity till you see exactly what you want.

Got it? Then, merge the bucket fill layer into the layer with your chosen line filter. And duplicate the resulting single, fused/merged layer immediately, so that you can go on to the next step.

With the colors set, it’s time to breathe life into this project. The magic happens when you begin to paint with light ― literally.

Look long and hard at the picture. See, in your mind’s eye, where the light is coming from. Can you see it? If you can see it, you can paint it, even with a mouse. The tools you’ll be painting with are dodge and burn. Dodge make areas of the painting brighter. Burn crushes the shadows or mid-tones, making them darker. You’re the artist: it’s up to you to paint with light and shadow, and create the picture. You have the outlines; you have the colors. Now ― paint. You don’t need a pen tablet for this; the work is not fine or detailed. Dodge and burn are more like “area weapons” than fine-scale tools. They’re best applied over significant portions of the image ―

Tip: go carefully, applying about 10% or 20% of dodge, and maybe 5% or 10% of burn on any one click. Build up the effect, little by little, and let the image take shape slowly rather than trying to do it all in one mouse click.

You’re the artist here. You’ll need to use your artist’s eye, your judgement, and create the image you saw in your mind’s eye. Photoshop can’t do this for you … this is where the art comes into its own. Experiment; find out what works; throw the failures into the bin and re-clone the previous layer for another attempt until you see just what you want. Note that dodge and burn can both be applied to highlights, midtones and shadows, but you really can’t dodge crushed shadows, and you also can’t actually burn featureless highlight areas to any great effect. Remember, zones close to 1 and 255 have no information in them, and it’s impossible to add information that isn’t there. So be careful how far you go with dodge and burn, because when the information is gone ― when you’ve brightened or darkened too far ― you won’t be able to add it back in. Go gradually, slowly, and build up. (If you do go too far and erase too much information, there’s a trick for getting it back … you can always paste the dead areas over from a previous layer, merge them into the new dodged, burned layer, and dodge/burn them juuust enough to fill in those dead zones. It’s tricky, but it does work.)

At this point, you might decide that you’re finished. If the painting looks done to you, that’s it. If it meets your original vision, you’re done. That’s art, and the call is yours to make.

For myself … I decided I wanted more texture in it. If I were painting by hand with the pen tablet, this is the time when I would start to doodle in color(s) and put in all the fiddly bits, the detail, or pseudo-detail that seduces the eye into thinking you’re looking at foliage when, in fact, you’re not. But we’re taking shortcuts here ― allowing for the fact that you might not have a pen tablet, or else you just don’t have an hour to two to spend faffing around, doodling. For whatever reason, you can take the shortcut.

You can apply another filter, and pour on more texture, save yourself the time and work. This is nothing that you can’t do by hand ― and it’s a lot of fun killing a whole afternoon or evening with this kind of detail doodling. But if you only have ten minutes … Photoshop to the rescue again.

The filter I chose was Poster Edges, with the settings dialed way down, so the filter didn’t ruin the picture….

And last of all, I added lens flare. Go to the very brightest area of the image ― where the sun would be, striking through the morning mist. Go to Filter > Render > Lens flare, and pick one of three types of lens flare. Position the hotspot. Dial the intensity up or down till it looks right to you. This was the point where I decided the picture was just what I wanted. I did try saturating the colors, but decided this “overcooked” it, and undid the effect. I could have gone on messing about … adding birds and bokeh and dust motes with ABR brushes, but you really can overcook a picture. One of the key things about art is knowing when to stop … when to sign off on the picture, and call it good!

And the result, in closeup, so you can actually see it on the screen here? Worth a quick look before I call this post done:

...and that's the point where I signed off on it and called it done! 

Creating character body morphs in DAZ Studio 3

As promised, here's the companion feature -- or continuation of the feature -- where we created a new, unique Michael 4 character, starting with the "doll" right out of the box. In the last segment we created the face from scratch. In this one, we're going to work on the body, and we'll wind up with the stunning creature you see striking a pose above! In fact, this one is such a stunner I went on and did a male nude glamour shoot with him, which will be posted to the Exotic blog in about an hour. Stay tuned.

Okay, so let's get started! First, return to the file where you were working on the face and back the camera off so you can see the whole body...

This is the doll, just as it loads. The only thing it's had done to it is, the JM Alexander skinmap (hairy ... default) is loaded, and I put a pair of Speedoes on him for the sake of modesty. Also, on this blog I won't be talking about how to load up and configure the dangly bits. That would be inappropriate for a blog where the faint of heart are known to tread, albeit infrequentlly

The first thing to notice is that when you select the M4 model, the Morphs++ menu opens up on the right side of your screen. These morphs are the familiar slider bar controls that give you absolute control over virtually everything.

Let's be silly before we get clever. Lets' grab the "whole body" slider controls and drag them as far as they'll go, see what happens:

Instant Incredible Hulk. How did a shaved gorilla get in here? Thank gods for the UNDO icon! So when you've seen how far you can push everything, undo the whole thing and get back to the M4 doll. Now, start with the full-body morphs and adjust them delicately to get a nice balance of body morphology...

NOTE: all the screencaps in this tutorial are uploaded at 1:1 size, so you can click to open them and see exactly where I dragged the controls. You can read the percentages off the images, if you wanted to absolutely duplicate this bod!

Now, this is a bit more like it. He looks like an athlete, really toned. Now you can get into the minor adjustments. This is the point where the visual changes can be rather subtle. You might have to look twice to see the difference, but in the one below I've been working on the arms and shoulders...

This is closer to what I had in mind when I started, so it's time to turn him around and organize the kind of bubble butt that makes it into the magazines:


So far, so good. I "turned off" the Speedoes so you can see the effect. You have the morphs to change the size and shape of the glutes, the flex and tension on them. I also changed the width of the hips and the amount of "love handles" he's carrying. This is starting to look really good, so -- "turn on" the Speedoes and turn him back around:

Starting to look very nice indeed. Starting to get that "bombshell model" look about him ... so, on a whim I decided to change the skinmap. It's still JM Alexander, but I switched to the "shaved" option. Models are always depilated:

At this point, it's also worth rendering one and seeing what the overall effect is. I don't have any lights set in this render. I have gone into the Surfaces tab and turned the "lighting model" to MATTE on both the M4 and the Speedoes. If you don't turn the skinmap's lighting model to MATTE, almost all of them get white blotches. And when the skinmap's lighting model has been changed this way, you also have to change the lighting model on the costume, or it'll shine like patent leather, which looks unrealistic.

Now, when you set the lighting models to MATTE, and you don't have any lights set up on the character, your first render will probably look dull, even lifeless. Don't worry about it -- this will come back to life when you set up your lights.

And now it's time to get "up close and personal." I won't be working below the belt in this feature, for obvious reasons, but unless you're happy with the generic chest, it's time to work on that, and you will find that you have control of everything. And I mean, everything.

The results are good enough now to set up some photo lights and render a proper one. Pose the model, put a slightly provocative expression on that angular, unique face we created last time, and here you are...
Set the background to black, render this properly, and save the render. Now you're cooking, and this model is absolutely unique:

So lets' see what we can do with him, in the photographic sense. Add a prop and a couple extra lights, turn him around so we wind up with a render we can display here ... lose the Speedoes (again) and render this again. A quick tweak in GIMP or whatever, and ...

Jade, 23 November

How to use maps in DAZ Studio 3

Today it's all about surfaces, and the textures you can slap onto 'em, and how you get these effects. I've been chatting with a visitor about this, and it's actually pretty interesting, so let's recruit Michael 4 and have him model for us!

In honor of Halloween, just past, we've cast him in the part of a Goth. That's the Elven Prince hair set to ebony, and the Lockwood costume. The skinmap is Albane ... an albino. Eventually, as our hero looked less and less like a Goth, I changed out the eyes for rich green ones, but at the outset he's an albino goth, red eyes and all:

So those are the textures right out of the box, just as they load. It's not bad, but it's ... ordinary. Nothing wrong with it, but it's ho-hum. A bit boring. So let's just DUMP all the textures and start over:

What you want to do is go into the Surfaces tab, select each of the items one at a time. Click in the field under DIFFUSE , and you open the dialog. Look up in the top left hand corner of the big list that opens. See None? Click it. One by one, you make the textures go away, and you wind up with virgin white plastic. So far, so good.

What you just did was to turn OFF the diffuse maps, and this instantly tells you what a diffuse map is, and what it does. They're the maps that put the fabrics or visible textures onto objects. Turn 'em off, and you see the objects for what they are.

Just playing here, try adding silly things back onto the nekkid objects. Right where you clicked "none" to turn off the diffuse maps, you can also click "browse," and choose new textures. You'll be doing this a lot, so play around with this, get used to it with something silly:

The guy now looks like a small explosion in a garment factory! But by doing this, you really get a grasp on what the diffuse maps are. You might be wondering where all the textures come from. I made all the textures which are used in these renders. Just make them yourself, using parts of photos, or webpage backgrounds, which you can download free in loads of places. You know how to make seamless textures from tiles ... same thing here.

So now let's put some decent textures on the model and get to work:

This is better: a tie-dye fabric texture on the shirt and leather on the pants. If you look at this full-size through, it still looks plastic. Not enough has been done with it to stop the plain object underneath from showing through. So, what else can you do with this, to make it look more real? This is where your maps come in!

First, let's get rid of the background and lights, to make the renders go faster -- we'll put them back in at the end. NOTE: from here on down, they're just test renders -- fast and flat, to see the results...

So right here I've added one extra map ... the OPACITY map. This is just a black and white image (pure black, pure white -- in the days of yore we used to call this zip-a-tone). The white stays, the black goes transparent. You add this map in the Basic surfaces tab. See the dialog there? Click in it, and browse for your displacement map. Where do you get these maps? Honestly -- make your own! Get a black felt-tip pen and draw a pattern on white paper. Scan it in, save it, an then set it as the opacity map. You can also get clever with GIMP or Serif, or whatever you use, but the end result is the same: a pure b/w image saved as a JPG and applied to an object. The black parts go transparent.

Notice, you can also play around with the percentage of opacity. This is how you can turn leather into silk. In fact --

What I did here was take the opacity map OFF the shirt model, and just lower the percentage of opacity. The more you pull the slider, the more transparent it gets. Nice! So --

Basically, play with the controls and make the costume fade in and fade out, till you figure out how it works. While you're doing this you'll start to notice that the opacity map creates a great pattern through the fabric, but it can wind up looking exactly like a pattern printed on the fabric, which isn't the effect you want. This is because it's just a 2D effect applied to the 3D object. Now, you want to make the fabric look as if it's transparent in 3D...

You'll probably have to click on the image to see the effect, but ... we're getting there now. What I've done for this render is get rid of the old diffuse map on the pants, because it was just leather and it's one flat color. I put on a map with some swirls of color so it's much more interesting, and then I gave the poor man a little something to wear under the costume, because it's rapidly vanishing. Then, the important part.

To get a 3D effect using a 2D map, you have to apply a DISPLACEMENT map. As the name suggests, the map "displaces" the surface of the object. And here's the trick: apply the same map you used as the OPACITY map -- set it as the displacement map too, and don't forget to crank the slider up to 200% to really see the change.

Now, here's where it gets interesting -- and fun:

Click on this to see it full size and see the effect. By playing around with the displacement maps you can create the effect of texture in fabric. What I've done here is take the opacity map OFF the pants, and then apply a DISPLACEMENT map to look like fabric texture. Also, notice that your displacement map controls have positive and negative values which you can jiggle. You can make the pants foof up or get tighter ... and if you then decrease the opacity to make them transparent, you get foofie harem pants made of gossamer silk. Now we're cooking!

So now you can set the diffuse map to decide the color, det the opacity map to decide the pattern of the opacity, juggle the percentage of the opacity to decide how transparent something is ... set the displacement map to get a lovely 3D effect, and jiggle the positive and negative values to set the degree of foofiness. What's next? Because it's still looking "flat" here. Well...

Now it's time to pay some attention to the surface values. You have three basic colors you can set: DIFFUSE, GLOSSINESS and AMBIENT. Technically, "ambient" is the color something is when it's in the dark. Now, strictly speaking that's dumb, because things in the dark are only ever black, right? But in the wonderful world of 3D art, "ambient" means the color something is before a light shines on it. A...ha. Glossiness speaks for itself: when you shine a light on something, how glossy does it get, and what color does it gloss up?

You get to set not only the color you want it to gloss up, but also the degree of glossiness. Here's what they don't tell you: 100% is NO GLOSS, and 0% is shining like a searchlight! Makes sense, doesn't it? Not.

While you're thinking about gloss, you should also think about REFLECTIVITY. The two things are not the same, though they're similar. Reflectivity adds an amazing effect to floors, walls, surfaces, widows ... fabrics. Things like silk and organza have loads of reflectivity, so don't forget to set this, while you're working with the gloss. Also, see below -- lighting models. When you set the lightning model to metallic, reflectivity really starts to work for you.

Next thing you need to be aware of is how the surfaces interact when they lie on top of each other. When you're working with garments, the shirt is probably over the pants, or the skirt is over the tights. By the time you've got loads of different maps and surface values stacked onto both the shirt and the pants, sometimes you don't get the effect you expect, or think you should get:

I can't give you a clear reason for why the surfaces interact the way they do. It might be down to the shortcomings of the render engine! I'm still using the 3DLight engine, which is free and ships with DAZ Studio 3, because I don't have the money to buy myself a candy apple at this point in time. I drool over the Octane render engine, and Reality etc, but those require a 64 bit system, which I also don't have. So like most people, I'm making do. And if you are too, you need to be aware that surfaces can and do interact a bit strangely when they're all loaded with values and maps and then overalaid.

What can you do about it? You can jiggle the settings -- mainly the opacity settings. Work with one garment at a time. Turn them on and off till you get something close to the effect you wanted.

Then, when it's all coming right, turn on the costume ... turn on the backround ... set the lights, pose the figure, and you're alllmost ready to render up the final shots, which in fact I started with, waaaay above! But first --

One more thing you can do in the surfaces tab. If you decide to leave the costume off, you often find that in the final render Michael 4 will render with white blotches or streaks on his skin. This is very annoying, and it happens on many, many skinmaps, up to and including the elite ones, which cost a packet. The solution is to be found in the surfaces tab. You want to go into the LIGHTING MODEL, and turn it not to "skin" but to "matte." This will fix the problem:

Now, let's return to the full-quality render ... takes time, but --

There's only one thing I did to the costume for the display renders at the top of this post, that I haven't talked about here ... except for just touching on it in the last paragraph. It's the LIGHTING MODEL. You'll need to find this, in order to fix skinmap problems, and once you have found it, you can also apply lighting model effects to fabrics or anything else. This looks almost like the shirt is made of metal:

...and the trick is to set the lighting model to (!) metallic, and then crank the REFLECTIVITY up full blast -- then shine some lights on it.

That's all of it, nothing kept secret. It's a heck of a lot of fun to play with -- it'll take a little while to learn it, but as soon as the penny drops, it's actually easy. Trust me on this -- I puzzled all this out for myself earlier this year!

Jade, 3 November

***Posted by MK: my connection is intermittent, too slow for this. Seriously, guys, I've got dialup speeds. How are you expected to do anything these days, at 1990 dialup speeds?!!!